Mapping Africa's

Building a collection of controlled vocabularies for the Arches Database

By MAEASaM Project Team

MAEASaM Arches database contains over 45 thematic collections (thesauri) of controlled vocabularies covering c. 5100 unique terms relating to heritage and archaeological concepts across Africa. Several of these collections have been organised into structured hierarchies ranging from general concepts to more specific, regionally based, terms and where possible include both Arabic and French translations. In this contribution, the MAEASaM team reflect on some of the challenges of compiling a vocabulary collection for African sites and monuments and offer some of the ways in which these challenges have been addressed for the Arches database.

The use of terms to define and categorise the past including landscapes, buildings, monuments, sites, artefacts and chronological periods, to name a few, has a long tradition in Archaeology. Like many other disciplines, the naming of ‘things’ along formal systems of reference offer us the ability to order and make meaning out of the ‘things’ we wish to describe. The use, for example, of a list of controlled vocabularies in a digital database, provides a consistent framework for describing data including its retrieval and exploration (Smith 2021). By standardising concepts, it allows us to share information with one another using a common language that can be understood by all. There remains a caveat, however. Whether working with analogue or digital records, we recognise that the naming and classification of objects, places, sites, chronological periods, etc., are often subjectively defined, dependent on context, and can be non-transferrable across different regions, time periods, and linguistic groups.

Our work in developing the MAEASaM controlled vocabulary collections has offered us the opportunity to reflect on the current use of terms in the Archaeology discipline and how we might further the inclusion of diverse and, at times, conflicting concepts relating to heritage through the use of multilingual thesauri that have alternative terms and concept qualifiers.

Importance of building a vocabulary collection

The use of controlled vocabularies in a digital environment allows us to compare heterogeneous heritage information by offering a shared form of communication to describe data through specified terms and concepts. Such vocabularies can be structured as simple alphabetical lists or built into hierarchies containing complex relationships to express broader and narrower concepts. Without the use of controlled vocabularies, we might lose the ability to harmonise heterogeneous heritage information and to conduct comparative analyses using different search queries. For example, by specifying a particular chronological period or site type, we are able to retrieve records about sites that share the same or similar conceptual terms not only at a regional scale of analysis but also at national and/or transnational scales of inquiry (fig. 1).

Grappling with concept definitions and practical applications

We often encounter multiple challenges in harmonising complex heritage information. For example, the defining of different site types and functions, site features, and chronological periods are by no means straightforward. This is something that has been highlighted by McCoy (2020), particularly around the issues associated with defining a ‘site’ and how the conceptualisation of such remains theoretically and practically problematic especially when visualising and representing it within geospatial contexts. Chronological concepts are by no means an exception. For example, a period that has been defined by one archaeologist may be differently defined by another. A period may also contain different meanings in different geographic contexts and timespans (Rabinowitz et al. 2016).

In trying to grapple with these ambiguities, the team began comparing various online Thesauri (including the Getty Art & Architecture Thesaurus, FISH Terminologies, and EAMENA Database for site concepts and PeriodO and iDAI ChronOntology for relative chronologies) and mapped these concepts to one another in order to identify any differences and/or similarities in the way they that have been structured and defined. The mapping exercise revealed interesting results: 1. there currently exists a great deal of variability in how different online Thesauri go about structuring, categorising and defining unique terms and that the structuring of such Thesauri is dependent on the aims and objectives of those who are organising and collating them; and 2. many of the regionally specific terms relating to different site types, features, and temporal markers for the large majority of sub-Saharan African archaeology are largely absent from online Thesauri, making any further comparative analysis difficult (fig. 2).

In light of these findings, it was important to make the MAEASaM vocabulary collections as flexible as possible where in some instances alphabetical lists could be generated without making structured hierarchical relationships whilst in other cases allowing for more complex relationships to be established. Definitions also had to be adapted, where appropriate, in order to cater for concepts relating to sub-Saharan African archaeology which have been missing from predominant Euro-American definitions.

Building the MAEASaM vocabulary collection

The MAEASaM vocabulary collections were compiled from both known and existing Thesauri (Table 1) as well as from an extensive but non-exhaustive literature survey of past and current African archaeological textbooks and regionally specific journal publications. The Relative Chronology vocabulary collection was expanded from 100 unique terms to contain over 500 regionally specific terms through the consultation of over 300 publications in both French and English. Like many other online period thesauri or gazetteers, the aim of the MAEASaM Chronology Collection is not to resolve dating problems, but instead to flag alternative interpretations and debates. Where applicable concepts may carry a qualifier to distinguish similar or same terms (names) but with different definitions based on geographic and/or temporal context.

Table 1: Available online Thesauri and gazetteers consulted for compiling the MAEASaM controlled vocabularies.

Getty Art & Architecture Thesaurus® Online (Getty AAT)

Structured resources that can be used to improve access to information for art, architecture, and other material culture. [Link]

Forum on Information Standards in Heritage (FISH Terminologies)

Terminology used for recording archaeological techniques, objects, monuments, materials, heritage threats, etc. [Link]

EAMENA Project

Vocabulary Collections from the Endangered Archaeology in the Middle East and North (EAMENA) Project. [Link]

International Council on Monuments and Sites (ICOMOS)

Guidance on Heritage Impact Assessments for Cultural World Heritage Properties.[Link]

PeriodO

A gazetteer of scholarly definitions of historical, art-historical, and archaeological periods. It eases the task of linking among datasets that define periods differently. [Link]

iDAI.ChronOntology

A webservice that connects period terms to datings. [Link]

The use of qualifiers in some of our vocabulary

The use of a concept qualifier allows us to distinguish the broader context of a term or another significant defining characteristic of that term. In the MAEASaM Relative Chronology thesaurus qualifiers have been used to specify the same terms but with different geographic or temporal extents, and to distinguish a term by the ‘chronology type’ by which it has been defined. Where applicable, qualifiers have been represented by parentheses (_) at the end of a term. Instances of qualifiers include for example: Middle Stone Age (Southern Africa), Acheulean (Senegal), Letsibogo (ceramic), Khami Indo-Pacific series (glass bead), Reef-coral buildings (architecture), etc.

Expanding and accessing the vocabulary collections

Whilst the project’s vocabulary collections will undoubtedly continue to expand with the addition of more regionally specific terms in Phase 2 of the project, we hope that the vocabulary work developed thus far will contribute towards placing African archaeological concepts and terms within an open linked data environment and by emphasising the need to revisit and perhaps redefine our past and current uses of terms in the African Archaeological discipline. We are also looking forward to the release of Lingo for Arches version 8. Lingo is a management tool for shared controlled vocabularies to describe cultural heritage data and will be more accessible to end-users. The release of Arches Lingo is expected for early next year (2026).


Citations

  • McCoy, M. D. (2020). The Site Problem: A Critical Review of the Site Concept in Archaeology in the Digital Age. Journal of Field Archaeology, 45(sup1), S18–S26. [Link]
  • Rabinowitz, A., Shaw, R., Buchanan, S., Golden, P., & Kansa, E. (2016). Making sense of the ways we make sense of the past: The PeriodO project. Bulletin of the Institute of Classical Studies, 59(2), 42-55. [Link]
  • Smith, C. (2021). Controlled Vocabularies: Past, Present and Future of Subject Access. Cataloging & Classification Quarterly, 59(2–3), 186–202.[Link]